Nowadays the sediment becomes one significant problem to reservoir watershed and it is effect and related to reservoir operation system. As the research topic, an integrated sediment approach and impacts of climate change on reservoir sedimentation, there are three main parts demonstrated in this research that all parts are related together with sediment point of view. Annual sediment depositing volume in reservoir was estimated by general soil loss equation but the efficiency was not acceptable. The first part of this study shows that the efficiency is improved by using general soil loss equation with sediment transport model. The second part is about monitoring the sediment inflow to reservoir. The important parameter to operate the reservoir is turbidity concentration of flow into dam, in the second part the suspended sediment concentration was predicted by real time therefore the reservoir operation to release turbid flow will get more efficiency. For last part, in the next future year sediment yield and water resources on the study area were investigated by extrapolated temperature and rainfall data then the results will be useful for long term reservoir operation system. First part, the integrated sedimentation was used to model an annual depositing sediment volume in reservoir. Sediment system in watershed includes not only sediment yield but also sediment transportation along the rivers. In this study, the Geographic Information System (GIS) incorporated with sediment yield model can be assisted to enhance the evaluation estimation of soil erosion. Surface erosion on Managawa river basin is then computed with the Modified Universal Soil Loss Equation (MUSLE) and it is verified to reflect the hydrological processes to which it will be able to estimate soil losses. In the sediment transport routing module, total load equation is applied to carry sediment from soil surface erosion to deposit in Managawa dam. According to annual accumulation sediment volume data in Managawa reservoir during 1981 – 2004, the establish model and simulation results are satisfied. The efficiency of the Modified Universal Equation with sediment routing in rivers is more than the simple Modified Universal Equation. Second part, the real time suspended sediment concentration forecasting was used for monitoring the turbidity flow on the upstream of reservoir. The sediment flow into the reservoir is a factor for decision support in real time reservoir operation therefore the serious area of sediment erosion of Managawa river basin, Japan is monitored by suspended sediment gauge. The hourly suspended sediment concentration at Okumotani station; the upstream of Managawa reservoir, was monitored and estimated by the artificial neural network (ANN) model that the input data were rainfall data and its products. This artificial neural network (ANN) was calibrated and validated by using recently suspended sediment data on heavy rainfall events from December 2006 to January 2008. Choosing an appropriate neural network structure and providing field data to that network for training purpose are address by using a constructive back propagation algorithm. Rainfall and its products; the computed discharge from rainfall runoff model and rainfall intensity, were applied as inputs to neural network. It is demonstrated that the artificial neural network (ANN) is capable of modeling the hourly suspended sediment concentration with good accuracy and the neural network model has efficiency more than the multiple linear regression (MLR) model and the sediment rating curve (SRC) model. Last part, the effects of climate change on water resources and sediment yield were investigated by climate change scenarios which the main meteorological data were rainfall and temperature data. Historic trends of temperature and precipitation on Managawa river basin were detected by parametric and nonparametric tests. The daily mean temperature data from 1981 to 2008 at Ono station, Fukui prefecture was the representative of temperature on the study area. The hourly rainfall data from 1981 to 2008 were obtained by Managawa dam office processed with the reliability of data and weighted data. From monotonic and step trend tests, the temperature trend was found herein to follow a clear and steady trend every month. The average annual temperature exhibited an increasing trend with a magnitude 0.4 ºC per decade. Application of the Mann-Kendall and Mann-Whitney test for rainfall time series on Managawa river basin showed no step change and no monotonic trend in Managawa precipitation. The average annual precipitation exhibited a decreasing trend with a magnitude 52 mm per decade. The weather generating models both temperature and rainfall expressed the high efficiency for validation step. The generated weather series 2009 - 2060; temperature and precipitation height, for future climatic conditions can be inputted into the soil loss equation to investigate the change in sediment sources and extrapolated rainfall can be inputted to rainfall runoff model to investigate the change in runoff for future climate change condition. The sediment yield rate should be reduced because of the decrease in precipitation. / Kyoto University (京都大学) / 0048 / 新制・課程博士 / 博士(工学) / 甲第14934号 / 工博第3161号 / 新制||工||1474(附属図書館) / 27372 / UT51-2009-M848 / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 角 哲也, 教授 田村 武, 教授 藤田 正治 / 学位規則第4条第1項該当

In the ocean, light from the surface dissipates quickly leaving sound the only way to see at a distance. Different sediment types on the ocean floor and water properties like salinity, temperature, and ocean depth all change how sound travels across long distances. Hard sediment types, such as sand and bedrock, are highly reflective while softer sediment types, such as mud, are more absorptive and change the received sound upon arrival. Unfortunately, the vast majority of the ocean floor is not mapped and the expenses involved in creating such a map are far too great. Traditional signal processing methods in underwater acoustics attempt to localize sources and estimate seabed properties, but require a priori decisions and fall victim to ill conditioning and non-linear relationships between the unknowns and are computationally expensive. To address these problems, a deep learning method is proposed to distinguish between seabed types while also predicting source parameters such as source-receiver range from simulated training data. In this thesis, several studies are presented that explore the effectiveness of convolutional neural networks to make predictions from two types of sounds that propagated through the ocean: impulsive explosions and ship noise. These studies show that time-series signals and spectrograms contain sufficient information for deep learning, and additional preprocessing for feature extraction is not necessary. Training data considerations, such as randomness in the network weights and inclusion of representative variability are also explored. In all, this study shows that deep learning is a useful tool in underwater acoustics and has significant potential for seabed parameter estimation.